740 THE FLORAL STEM. 



always takes place, because very often the crossing is prevented from some cause 

 or another. The event of failure is also actually provided for; in case the crossing 

 of diUbrcnt plants does not succeed, cai-e is taken that in the second stage of 

 flowering the pollen should reach the stigmas of the neighbouring flowers of the 

 same plant. In most plants only when this plan also fails, and at the last 

 moment, so to speak, does the pollen developed in the stamens of a flower reach 

 the stigma of the same flower which hitherto has i-emained intact although placed 

 in the closest proximity. The wonderful and extremely complicated contrivances 

 which are met with for the attainment of this threefold aim will be considered 

 fully in the second volume; but they must here be mentioned cursorily, because 

 the peculiar grouping and the remarkable order of the opening of the flowers repre- 

 sent contrivances which render possible the crossing of neighbouring flowers, and 

 because the shape of the inflorescence can only be comprehended in connection with 

 these contrivances. 



In thousands of different species it can be seen that in the event of failure of 

 crossing between flowers of different plants, a cross-fertilization between neigh- 

 bouring flowers is brought about by elongations, shortenings, depressions, and 

 various other alterations of position, sometimes of the style, sometimes of the 

 stamens, of the floral receptacle, or of the flower-stalks. In the racemose inflores- 

 cences of Eremurus (a liliaceous plant) the long styles of the lower flowers, 

 which are directed towards the rachis, bend upwards, towards the end of the flower- 

 ing period, in order to obtain pollen from the younger flowers above; and the same 

 thing occurs in the floral fascicles of a Woodruff" (Asperula taurina), in which the 

 styles bend down laterally to the neighbouring flowers in order to come into contact 

 with their pollen-laden anthers. The stamens of the Wayfaring Tree ( Viburnum 

 lantana) curve down towards the neighbouring flowers so that the pollen, falling 

 from their anthers, must alight on the stigmas of these neighbouring flowers. The 

 same thing happens in Eacquetia, Ghoerophyllum hirsutum, Siler trilobwm,, and 

 various other umbelliferous plants. In these we find that the stamens of the 

 flowers in the centre of the umbel stretch out so far that their pollen-laden anthers 

 are situated above the stigmas of the neighbouring older flowers, already deprived 

 of stamens, at the periphery of the umbel. In Anthriscus sylvestris the younger 

 umbels are placed above the older so that the pollen falling from the former must 

 necessarily reach the latter standing below them. 



In numerous composites, especially in asters and Golden-rod {Aster and Solidago), 

 as well as in species of Cacalia, Senecio, and Arnica, the tubular florets are so 

 arranged in the centre of the capitulum that the pollen expelled from the younger, 

 inner flowers necessarily falls on the stigmas of the adjacent, outer flowers without 

 the aid of any special elongation or curvature. In those composites, on the other 

 hand, of which the Chamomile {Matricaria cliamomilla) may be taken as a type, 

 the stigmas of the older, peripheral flowers are brought under the pollen falling 

 from the inner, younger flowers by an elongation of the arched or conical rachis, 

 and by the slight raising or displacement of the flowers of the capitulum so pro- 



